Spaces:
Running
Running
File size: 11,383 Bytes
64a008c b7dddbe 5a01a63 64a008c b7dddbe 64a008c b7dddbe 64a008c ea0e2eb 64a008c ea0e2eb 64a008c ea0e2eb 64a008c ea0e2eb 64a008c faefb1f 64a008c | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 | from __future__ import annotations
import json
from typing import Annotated, Any, Dict, List, Optional
from fastapi import APIRouter, File, Form, HTTPException, Request, UploadFile
from pydantic import BaseModel, ValidationError
from app.api.deps import get_redis_scripts
from app.config import get_settings
from app.services.chat_service import chat_completion
from app.services.csv_analysis_service import (
execute_csv_chat_blocks,
get_dataset_info,
)
from app.services.prompts import get_csv_system_prompt
from app.utils.json_utils import extract_json_blocks
class _AnalyzeBlock(BaseModel):
description: str = ""
python_code: str = ""
class _VisualizationBlock(BaseModel):
description: str = ""
python_code: str = ""
class _AIResponse(BaseModel):
analyze: List[_AnalyzeBlock] = []
visualization: List[_VisualizationBlock] = []
message: str = ""
router = APIRouter()
_settings = get_settings()
_MAX_UPLOAD_BYTES = _settings.max_upload_bytes
@router.post(
"/csv/info",
summary="Get metadata for up to 10 CSV files (upload or URL)",
)
async def get_csv_info(
files: Annotated[Optional[List[UploadFile]], File(description="CSV files to inspect (max 10 total with URLs)")] = None,
urls: Annotated[Optional[str], Form(description="JSON array of file URLs (max 10 total with files)")] = None,
):
parsed_urls: List[str] = []
if urls:
try:
parsed_urls = json.loads(urls)
if not isinstance(parsed_urls, list) or not all(isinstance(u, str) for u in parsed_urls):
raise ValueError("urls must be a JSON array of strings")
except (json.JSONDecodeError, ValueError) as exc:
raise HTTPException(status_code=400, detail=str(exc))
file_count = len(files) if files else 0
url_count = len(parsed_urls)
total = file_count + url_count
if total == 0:
raise HTTPException(status_code=400, detail="Provide at least one file or URL")
if total > 10:
raise HTTPException(status_code=400, detail=f"Maximum 10 sources allowed (got {total})")
results: List[dict] = []
if files:
for f in files:
try:
data = await f.read()
except Exception as exc:
results.append({"source": getattr(f, "filename", "unknown"), "success": False, "error": f"Read error: {exc}"})
continue
if len(data) > _MAX_UPLOAD_BYTES:
results.append({"source": f.filename or "unknown", "success": False, "error": f"File exceeds {_settings.max_upload_mb} MB limit"})
continue
if not data:
results.append({"source": f.filename or "unknown", "success": False, "error": "Empty file"})
continue
try:
meta = await get_dataset_info(data)
meta["source"] = f.filename or "upload"
results.append(meta)
except Exception as exc:
results.append({"source": f.filename or "upload", "success": False, "error": str(exc)})
for url in parsed_urls:
if not url.startswith(("http://", "https://")):
results.append({"source": url, "success": False, "error": "Only http/https URLs are supported"})
continue
try:
meta = await get_dataset_info(url)
meta["source"] = url
results.append(meta)
except Exception as exc:
results.append({"source": url, "success": False, "error": str(exc)})
return {
"success": True,
"total": total,
"succeeded": sum(1 for r in results if r.get("success")),
"failed": sum(1 for r in results if not r.get("success")),
"results": results,
}
# @router.post(
# "/csv/analyze",
# summary="Execute Python analysis code against a CSV file (upload or URL)",
# )
# async def analyze_csv(
# file: Annotated[Optional[UploadFile], File(description="CSV file to analyze")] = None,
# url: Annotated[Optional[str], Form(description="URL to a CSV file")] = None,
# code: str = Form(..., description="Python code to execute (df pre-loaded with CSV data)"),
# token: str = Depends(require_auth),
# ):
# if not file and not url:
# raise HTTPException(status_code=400, detail="Provide either a file or a URL")
# if file and url:
# raise HTTPException(status_code=400, detail="Provide either a file or a URL, not both")
# if file:
# data = await file.read()
# if len(data) > _MAX_UPLOAD_BYTES:
# raise HTTPException(status_code=413, detail=f"File exceeds {_settings.max_upload_mb} MB limit")
# if not data:
# raise HTTPException(status_code=400, detail="Empty file")
# result = await analyze_csv_dataset(data, code)
# else:
# result = await analyze_csv_dataset(url, code)
# return result
# @router.post(
# "/csv/chart",
# summary="Generate a chart from CSV data and return as base64 PNG (upload or URL)",
# )
# async def chart_csv(
# file: Annotated[Optional[UploadFile], File(description="CSV file for chart generation")] = None,
# url: Annotated[Optional[str], Form(description="URL to a CSV file")] = None,
# code: str = Form(..., description="Python chart code (df pre-loaded, use matplotlib/seaborn)"),
# token: str = Depends(require_auth),
# ):
# if not file and not url:
# raise HTTPException(status_code=400, detail="Provide either a file or a URL")
# if file and url:
# raise HTTPException(status_code=400, detail="Provide either a file or a URL, not both")
# if file:
# data = await file.read()
# if len(data) > _MAX_UPLOAD_BYTES:
# raise HTTPException(status_code=413, detail=f"File exceeds {_settings.max_upload_mb} MB limit")
# if not data:
# raise HTTPException(status_code=400, detail="Empty file")
# result = await create_csv_chart(data, code)
# else:
# result = await create_csv_chart(url, code)
# return result
@router.post(
"/csv/chat",
summary="Chat with AI about a CSV file — returns analysis + chart code results",
)
async def csv_chat(
request: Request,
file: Annotated[Optional[UploadFile], File(description="CSV file to analyze")] = None,
):
content_type = request.headers.get("content-type", "")
url: Optional[str] = None
query: Optional[str] = None
if "application/json" in content_type:
try:
body = await request.json()
url = body.get("url")
query = body.get("query")
except Exception as exc:
raise HTTPException(status_code=400, detail=f"Invalid JSON body: {exc}")
else:
form = await request.form()
url = form.get("url")
query = form.get("query")
has_file = file is not None
has_url = bool(url)
if not has_file and not has_url:
raise HTTPException(status_code=400, detail="Provide either a file or a URL")
if has_file and has_url:
raise HTTPException(status_code=400, detail="Provide either a file or a URL, not both")
if has_file:
data = await file.read()
if len(data) > _MAX_UPLOAD_BYTES:
raise HTTPException(status_code=413, detail=f"File exceeds {_settings.max_upload_mb} MB limit")
if not data:
raise HTTPException(status_code=400, detail="Empty file")
source: Any = data
else:
source = url
if not query:
raise HTTPException(status_code=400, detail="Query is required")
metadata = await get_dataset_info(source)
system_prompt = get_csv_system_prompt(metadata)
messages = [
{"role": "system", "content": system_prompt},
{"role": "user", "content": query},
]
redis, scripts = get_redis_scripts(request)
try:
ai_response = await chat_completion(
messages=messages,
response_format={"type": "json_object"},
max_tokens=12000,
redis=redis,
scripts=scripts,
)
except RuntimeError as e:
raise HTTPException(status_code=502, detail=str(e))
parsed = ai_response.get("parsed")
if not parsed:
choices = ai_response.get("choices", [])
content = choices[0].get("message", {}).get("content", "") if choices else ""
blocks = extract_json_blocks(content)
if blocks:
parsed = blocks[0]
else:
try:
parsed = json.loads(content)
except (json.JSONDecodeError, TypeError):
pass
if not isinstance(parsed, dict):
return {
"success": False,
"message": None,
"analyze": [],
"visualizations": [],
"error": "AI response was not valid JSON",
}
try:
ai_data = _AIResponse(**parsed)
except ValidationError as exc:
return {
"success": False,
"message": None,
"analyze": [],
"visualizations": [],
"error": f"AI response failed schema validation: {exc}",
}
message_text = ai_data.message
has_content = bool(message_text.strip()) if message_text else False
analyze_blocks_raw = [b.model_dump() for b in ai_data.analyze]
viz_blocks_raw = [b.model_dump() for b in ai_data.visualization]
exec_result = await execute_csv_chat_blocks(
source=source,
analyze_blocks=analyze_blocks_raw,
viz_blocks=viz_blocks_raw,
)
if not exec_result["success"]:
return {
"success": False,
"message": ai_data.message if has_content else None,
"analyze": [],
"visualizations": [],
"error": exec_result.get("error", "Code execution failed"),
}
results = exec_result.get("results", {})
raw_analyze = results.get("analyze", [])
raw_visualizations = results.get("visualization", [])
analyze_results: List[Dict[str, Any]] = []
for i, block in enumerate(ai_data.analyze):
raw = raw_analyze[i] if i < len(raw_analyze) else {}
code = block.python_code.strip()
if not code:
continue
analyze_results.append({
"description": block.description,
"code": code,
"success": raw.get("success", False),
"output": raw.get("output", ""),
"error": raw.get("error"),
"execution_time_ms": exec_result["execution_time_ms"],
})
viz_results: List[Dict[str, Any]] = []
for i, block in enumerate(ai_data.visualization):
raw = raw_visualizations[i] if i < len(raw_visualizations) else {}
code = block.python_code.strip()
if not code:
continue
viz_results.append({
"description": block.description,
"code": code,
"success": raw.get("success", False),
"image_base64": raw.get("image_base64"),
"error": raw.get("error"),
"execution_time_ms": exec_result["execution_time_ms"],
})
return {
"success": True,
"message": ai_data.message if has_content else None,
"analyze": analyze_results,
"visualizations": viz_results,
"error": None,
}
|